Activation of Persulfate by Irradiated Magnetite ... - ACS Publications

Dec 10, 2014 - *Tel. +33-(0)2-23238027. Fax +33-(0)2-23238120. E-mail: ... in the presence of persulfate (S2O82–) under UVA irradiation. ... Environ...
0 downloads 0 Views 975KB Size
Subscriber access provided by UNIV OF WATERLOO

Article

Activation of persulfate by irradiated magnetite: Implications for the degradation of phenol under heterogeneous photo Fenton-like conditions Paola Avetta, Alessia Pensato, Marco Minella, Mery Malandrino, Valter Maurino, Claudio Minero, Khalil Hanna, and Davide Vione Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/es503741d • Publication Date (Web): 10 Dec 2014 Downloaded from http://pubs.acs.org on December 16, 2014

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23

Environmental Science & Technology

1 2 3

Activation of persulfate by irradiated magnetite: Implications for the degradation of phenol under heterogeneous photo Fenton-like conditions

4 5 6 7

Paola Avetta,a Alessia Pensato,a Marco Minella,a Mery Malandrino,a Valter Maurino,a Claudio Minero,a Khalil Hanna,b,* Davide Vione a,*

8 9

a

10 11 12 13 14

b

Dipartimento di Chimica, Università di Torino, Via Pietro Giuria 5, 10125 Torino, Italy. Ecole Nationale Supérieure de Chimie de Rennes, CNRS, UMR 6226, 11 Allée de Beaulieu, CS 50837, 35708 Rennes Cedex 7, France.

* Address correspondence to either author. Tel: +39-011-6705296; Fax +39-011-6705242 (D. Vione); Tel. +33-(0)2-23238027; Fax +33-(0)2-23238120 (K. Hanna). E-mail: [email protected] (D. Vione), [email protected] (K. Hanna).

15 16 17

Abstract

18 19 20 21

We show that phenol can be effectively degraded by magnetite in the presence of persulfate (S2O82−) under UVA irradiation. The process involves the radical SO4−•, formed from S2O82− in the presence of FeII. Although magnetite naturally contains FeII, the air-exposed oxide surface is fully oxidized to FeIII and irradiation is required to produce FeII. The magnetite +

22 23 24

S2O82− system was superior to the corresponding magnetite + H2O2 one in the presence of radical scavengers and in a natural water matrix, but it induced phenol mineralization in ultrapure water to a lesser extent. The leaching of Fe from the oxide surface was very limited, and

25 26 27 28

much below the wastewater discharge limits. The reasonable performance of the magnetite/persulfate system in a natural water matrix and the low levels of dissolved Fe are potentially important for the removal of organic contaminants in wastewater.

29 30

ACS Paragon Plus Environment

Environmental Science & Technology

31

Page 2 of 23

Introduction

32 33

The sulfate radical (SO4−•) is a strong one-electron oxidant.1 Its properties, including the

34 35 36 37

reactivity with organic compounds, have been determined in early studies where SO4−• was generated by radiolysis of persulfate solutions.2,3 The activation of persulfate (S2O82−/SO42−, E° = 2.01 V vs. NHE) to generate SO4−• (SO4−•/SO42−, E° = 2.6 V vs. NHE) was recently proposed to remove different kinds of organic pollutants in contaminated waters and soils.4-6 Persulfate activation to SO4−• can be initiated under photochemical, thermal (also with metal

38 39 40 41 42 43 44

catalysts) or chemical conditions, and SO4−• can rapidly react with several organic compounds with second-order rate constants in the range of 107-1010 M−1 s−1.7 Iron (often as dissolved Fe2+) is a commonly used transition metal for the chemical activation of persulfate:8-10 FeII + S2O82− → FeIII + SO42− + SO4−•

The •OH radical can also be produced in the presence of SO4−•. Reaction (2) takes place under basic conditions:9

45

SO4−• + OH− → SO42− + •OH

46

and reaction (3) at all pH values:

47 48 49 50 51 52

(1)

SO4−• + H2O → •OH + HSO4−

(2)

(3)

Liang et al. 4 suggested that SO4−• predominates under acidic conditions and •OH under alkaline ones. Both •OH and SO4−• could attack target contaminants, but they can also react with the persulfate anion as follows:7 OH + S2O82−→ HSO4− + SO4−• + ½O2 (or •OH + S2O82−→ S2O8−• + OH−) SO4−• + S2O82− → S2O8−• + SO42−



(4) (5)

53

Other competitive or scavenging reactions may occur (although reaction (7) would be minor

54

in the presence of organic compounds):7,9,11

55 56

FeII + SO4−• → FeIII + SO42− 2 SO4−• → S2O82−

(6) (7)

57 58 59

Because of similarities between the activation of persulfate by FeII and the Fenton reaction, the persulfate process is usually named Fenton-like. The main drawback of dissolved FeII (Fe2+) is the rapid oxidation and precipitation as

60 61 62

FeIII, which inactivates ferrous iron. Chelating agents have been used to maintain FeIII in solution, where it can be more easily recycled to FeII.8,12 In alternative, the UV irradiation of FeIII species is an interesting FeII source.13,14 Another possibility is the use of minerals such as

ACS Paragon Plus Environment

2

Page 3 of 23

Environmental Science & Technology

63 64

ferrihydrite, goethite, manganese oxide and clay (including modified clays),15,16 which can activate persulfate and induce the degradation of e.g. trichloroethylene or diesel

65

compounds.5,17 Magnetite (Fe3O4) is a good agent to activate persulfate in the remediation of

66 67

PAH-contaminated soil.18 Fe3O4 is one of the most stable mixed-valence oxides (FeII-FeIII) at ambient temperature, and it is the most abundant in natural settings. Moreover, it can be

68 69 70 71

synthesized in the laboratory by various biotic and abiotic pathways. Depending on the source or the synthesis method, magnetites with different composition, crystal habits, morphologies and surface properties can be produced.19 In addition to the magnetic properties that facilitate phase separation, the natural

72 73 74 75 76

presence of FeII in magnetite has understandably attracted attention in the field of heterogeneous Fenton and Fenton-like reactions.20,21 Moreover, we have recently found that the Fenton reactivity of magnetite can be considerably enhanced under irradiation.22 The traditional photo-Fenton process is exemplified in reactions (8-13), according to the proposed Haber-Weiss mechanism, and it is particularly useful with sunlight as radiation source.23 Note

77 78 79

that L is an intentionally added or naturally occurring organic ligand (e.g. citrate, oxalate or ligand moieties of dissolved organic matter). The •OH yield of reaction (13) is not quantitative, and additional oxidants (e.g. ferryl, FeO2+) could also be formed in the

80 81 82 83

process:24,25 FeIIIOH2+ + hν → FeII + •OH FeII + O2 → FeIII + O2•– (or FeII + H+ + O2 → FeIII + HO2•)

(8) (9)

84 85 86

HO2• + O2•– + H+ → H2O2 + O2 FeIII + O2•– → FeII + O2 FeIII-L + hν → FeII + L+•

(10) (11) (12)

87

FeII + H2O2 → FeIII + •OH + OH−

(13)

88 89

The use of magnetite to activate persulfate under irradiation has never been tested to date. Therefore, the goal of the present work is, first of all, to provide the very first evidence to our

90 91 92

knowledge of the feasibility of the process for the degradation of organic compounds in both synthetic and natural water matrices. Considering that magnetite exists as micrometric and nanometric particles in many natural and engineered systems, four kinds of synthetic

93 94 95 96 97

magnetites with different particle size and stoichiometry were tested in dark and irradiation experiments. Insights are given into the possible reactive species involved in the degradation pathways, and the systems magnetite + H2O2 + UV and magnetite + S2O82− + UV are compared for the degradation and mineralization of phenol.

ACS Paragon Plus Environment

3

Environmental Science & Technology

98

Page 4 of 23

Experimental section

99 100 101 102

Reagents and materials

103 104 105

hydrogen phthalate (99%), KI (98%), NaHCO3 (99%), NaH2PO4 (99%), Na2HPO4 (98%), 4aminoantipyrine (98%), K2CO3 (99%), H3PO4 (85%), HClO4 (70%), magnetite (97%), methanol (gradient grade) and horseradish peroxidase (type VI) were purchased from Sigma-

106

Aldrich, Na2S2O8 (99%) from Fluka, magnetite (98%) from Prolabo, H2O2 (35%) from VWR

107

International, argon (research grade) and zero-grade air from Sapio (Monza, Italy). All

108 109 110 111

reagents were used as received, without further purification. The aqueous solutions were prepared by using Milli-Q water (TOC < 2 ppb, resistivity ≥18.2 mΩ cm). Concerning the used magnetite specimens, two (S1 and S2) were synthesized by FeIIinduced mineralogical transformation of 2-line ferrihydrite and lepidocrocite (γ-FeOOH), respectively, as explained in detail in previous works.22,26 Magnetite S3 was purchased from

112 113 114 115 116 117 118 119 120 121

Phenol (purity grade 99%), 1,10-phenanthroline (99%), 2-propanol (99.9%), potassium

Prolabo and S4 from Aldrich. Sample characterization is explained in the Supporting Information (hereafter SI). Table 1 reports particle size, surface area, the pH values of the point of zero charge (pHPZC, estimated from potentiometric titration), and the FeII/FeIII ratios of the oxides. The latter were determined by chemical analysis after acid dissolution, as suggested by Gorski et al. 27 who reported an excellent agreement with Mössbauer spectroscopy. The ratio between ferrous and ferric ions (x = FeII/FeIII) was lower than 0.5, thus all the investigated samples may be considered as non-stoichiometric or partially oxidized magnetite.27,28 The most oxidized samples (lowest x values) were S1 and S4.26,27 S4 is a

122 123 124

commercial sample, but S1 could still contain precursor ferrihydrite as residual from the synthesis method. This is supported by TEM images (Figure S1-SI) and by the high PZC value of S1 (close to that of ferrihydrite, i.e. ~8.2,19 Table 1). Abiotic synthesis methods (as

125 126 127

for samples S1 and S2) generally fail to generate stoichiometric magnetite,27,28 probably because the washing step(s) easily cause Fe2+ dissolution/oxidation.27,29 Characterization of S1-S4 by X-ray powder diffraction (XRD) only gave the main magnetite diffraction peaks,

128 129 130 131 132

while other crystallographic phases were lower than the XRD limit of detection (around 5%). The PZC of S2-S4 (Table 1) is close to literature reports for magnetite.19 An expected inverse proportionality was observed between BET surface area and particle size. Significant aggregation of magnetite particles in aqueous suspension is suggested by the comparison of the particle size data estimated from TEM images (Table 1) with the hydrodynamic radii from

133 134 135

Dynamic Light Scattering (DLS, Figure S1-SI). Further details about the characterization methods, including XRD, TEM-EDAX (Transmission Electron Microscopy-Energy Dispersive X-ray spectrometry), multipoint N2–BET and DLS analyses are given in the SI.

ACS Paragon Plus Environment

4

Page 5 of 23

Environmental Science & Technology

136 137 138

Irradiation experiments

139 140 141 142

Magnetite stock suspensions at 1.0 g L−1 loading were prepared by ultrasonication (Branson 2200 ultrasonic bath, 40 kHz). The suspensions for irradiation (50 mL total volume in a beaker) were prepared by diluting stock suspensions to 0.2 g L−1. Phenol, H2O2 and persulfate were added from separate stock solutions. The natural pH value of the system was ∼5,

143 144

measured (and periodically monitored as irradiation progressed) with a combined glass electrode connected to a Metrohm 713 pH meter. Irradiation took place under a 20 W Philips

145

TL 09N UVA lamp, with emission maximum at 365 nm. The lamp irradiance on top of the

146 147 148 149 150 151 152 153 154 155 156

suspensions was 18 W m−2 in the 295-400 nm range, measured with a CO.FO.ME.GRA. (Milan, Italy) power meter. The suspension temperature was ∼30°C. A picture showing the experimental set-up is provided in the supplementary material of previous work.22 Irradiation experiments were carried out in duplicate, the results averaged and the error bars reported on the relevant plots. To monitor the time evolution of phenol, aliquots of ∼1.5 mL were withdrawn from the irradiated suspension at selected time intervals and immediately filtered on Millipore HV syringe filters (Teflon, 0.45 µm pore diameter). A precisely measured volume (0.7 mL) of the filtered solution was immediately added to an HPLC vial, containing 0.7 mL methanol to quench the reaction. Analysis within the shortest possible time was carried out by highperformance liquid chromatography coupled to diode array detection (HPLC-DAD). In some

157 158 159

experiments, larger aliquots (5 mL) were withdrawn to monitor persulfate and H2O2 (UV-Vis spectrophotometry), sulfate and carboxylic acids (ion chromatography), as well as total dissolved Fe (inductively coupled plasma). Further analytical details are provided as SI.

160 161 162

The analysis of total organic carbon (TOC) was carried out on filtered-only sample aliquots, without methanol addition, by using a Shimadzu TOC-VCSH Total Organic Carbon Analyzer, equipped with an ASI-V autosampler and fed with zero-grade air. The TOC was

163 164 165

determined as total carbon minus inorganic carbon. Natural water was sampled in summer 2013 from the surface of the eutrophic Lake Candia (near Torino, NW Italy) and vacuum filtered upon arrival to the laboratory (cellulose acetate filters, 0.45 µm, Sartorius). Sampled water had the following features: 7±1 mg L−1 Ca, 18±2

166 167 168 169 170

mg L−1 Mg, 15±1 µg L−1 nitrate, 6.9±1.0 mg C L−1 TOC, 12±1 mg C L−1 inorganic carbon, pH 8.3.

ACS Paragon Plus Environment

5

Environmental Science & Technology

171

Page 6 of 23

Results and Discussion

172 173

Preliminary runs were carried out to look for the main factors involved in phenol degradation

174

and to check for the optimal experimental conditions. The degradation of phenol with magnetite usually required the contemporary presence of radiation and an oxidant (H2O2 or

175 176 177 178

persulfate). The optimal magnetite loading was 0.2 g L−1, most likely because higher loadings would cause excessive scattering of radiation by the suspension.22 The need to apply radiation could be avoided at elevated persulfate concentration, and

179

effective transformation of 0.1 mM phenol took place with 50 mM S2O82− in the dark. In this

180

case the suspension pH dropped from ∼5 to ∼3 in less than 2 h, followed by very limited

181 182 183 184

further changes up to at least 24 h. Phenol degradation in the dark with 50 mM S2O82− also took place without magnetite, thereby suggesting that persulfate can directly react with phenol. The reaction would cause an initial pH decrease, as a consequence of the oxidation of the organic substrate with production of sulfuric acid (see for instance reaction (14), where

185 186 187

HPhOH represents phenol):30

188 189 190 191 192 193 194 195 196 197 198 199 200

HPhOH + S2O82− + H2O → HOPhOH + 2 SO42− + 2 H+

(14)

Due to contact with air, the surface layer of magnetite is fully oxidized to FeIII.27 The initial acidification would cause dissolution of surface FeIII, letting sub-surface FeII in contact with the aqueous phase. Persulfate activation and further acidification would proceed as follows:30 Fe2+ + S2O82− → Fe3+ + SO4−• + SO42− Fe3+ + 3 H2O → Fe(OH)3 + 3 H+

(15) (16)

The formation of poorly soluble FeIII species would limit the concentration of dissolved Fe, which was around 0.1 mg L−1. Moreover, acidification due to reactions (14-16) would induce an additional pathway of persulfate activation (reactions (17,18), with no net pH impact):30,31 S2O82− + H+ → HS2O8− HS2O8− → 2 SO4−• + H+

(17)

201 202 203 204

(18)

Differently from 50 mM persulfate, little to no degradation of phenol was observed with 0.5 mM S2O82− alone (without magnetite), either in the dark or under irradiation. In this case, both

205 206 207 208

magnetite and UVA radiation were required for phenol degradation. Further experiments with the irradiated systems suggested that no transformation was operational with phenol + UVA or with phenol + magnetite + UVA. The blank runs also showed negligible adsorption of phenol on magnetite.

ACS Paragon Plus Environment

6

Page 7 of 23

Environmental Science & Technology

209 210 211

Phenol degradation by magnetite + S2O82− + UVA: effect of the Fe3O4 specimen

212 213 214 215

Figure 1 shows the time trend of 0.1 mM phenol in the presence of 0.5 mM S2O82− under UVA irradiation, with no magnetite and in the presence of each of the studied magnetite specimens (S1 to S4, at a loading of 0.2 g L−1 each). Compared to the experiment without magnetite, all the S1-S4 samples enhanced phenol transformation. The enhancement was

216 217

highest for S2 and S3. The elevated photoactivity of S2 and S3, compared to S1 or S4, cannot be accounted for

218

by differences in particle size or surface area (see Table 1). However, S2 and S3 are the

219 220 221 222

studied magnetite specimens with the highest structural content of FeII (Table 1). The link between FeII content and photo-Fenton or photo-Fenton-like reactivity is only apparently straightforward (FeII is the most reactive Fe species towards peroxides),32 because the studied magnetites were unreactive in the dark with 0.5 mM S2O82−. The insignificant dark reactivity is in clear contrast with a simple hypothesis concerning the availability of FeII at the magnetite

223 224 225

surface, which would be higher for S2 and S3 compared to S1 and S4. The need to irradiate the system suggests that FeIII is the primary actor of magnetite photoreactivity, because FeII

230 231 232

species are often photochemically inactive.14 Radiation absorption by FeIII compounds (including the hydroxocomplexes) is linked to a ligand-to-metal charge transfer and the photolysis induces FeIII reduction to FeII, together with ligand oxidation (see reaction 8).33,34 To account for the observed phenomena one should consider that, although bulk magnetite contains FeII, its air-exposed surface is substantially or totally oxidized to FeIII. Indeed, FeII in the outer regions reacts with atmospheric O2 and forms a thin layer of epitaxial maghemite.27,35 The oxidation process occurs to a depth of approximately 3 nm and depends

233 234 235

on particle size and lattice order.36 Surface oxidation was confirmed by XPS analysis of the investigated samples (see Figure S2-SI), and it is corroborated by the relatively low values of FeII/FeIII (Table 1). FeIII photoreduction is thus required to produce FeII that activates

236 237 238

persulfate. Note that magnetite stoichiometry (FeII/FeIII) may influence its overall reactivity, including sorption capacity and reduction potential.29 In the presence of the photoactive magnetites S2 and S3, the suspension pH dropped from

239 240 241 242 243

∼5 to ∼4 upon irradiation. Considering that the direct reaction (14) between phenol and persulfate can be excluded (no degradation of 0.1 mM phenol was observed with 0.5 mM S2O82−), the pH drop would be accounted for by reactions (15,16). In these systems, dissolved Fe was detectable but quite low (< 0.1 mg L−1). Fe leaching is a potential problem because of the limits for wastewater discharge, but Fe levels below 0.1 mg L−1 are safe from this point of

244 245

view.15 The detected concentration of total Fe also places an upper limit for the possible scavenging of SO4−• by dissolved Fe2+, based on the reported reaction rate constants:7

226 227 228 229

ACS Paragon Plus Environment

7

Environmental Science & Technology

246 247

Page 8 of 23

compared to phenol, Fe2+ would scavenge